Many online day-traders are amateurs that lose more money than they make. I want to start a website posing as an online stock broker like E-trade. For the most part when customers on my site buy a stock, I just tell them they bought it but I keep the money. When they sell it, I of course credit the money back to their account. If the stock went down, I make money. Of course sometimes the stock may go up, in which case I take a loss. But overall in the long term I “make” money for myself that otherwise would have been “lost” forever on the stock market. The end result to the customer is exactly the same.

I would have a program that analyzes the buying behavior of the gamblers (er, I mean investors). It would flag those that are not really day traders (if they actually buy investments and hold them long term, or if they consistantly make money even through day trades). For any such customers that are flagged I really would have to start buying the actual stocks they are ordering, to protect myself in case the stock values go up.
-- ServoMan314, Jun 16 2008

Contract for Difference (CFD) http://en.wikipedia...ract_for_difference
a contract between two parties, typically described as "buyer" and "seller", stipulating that the seller will pay to the buyer the difference between the current value of an asset and its value at contract time [ed, Nov 05 2008]

The king of the Bucket Shops http://en.wikipedia...of_a_Stock_Operator
From when online meant a telegraph. Note the Marks knew the bucket shop operater was holding their bets from the start. Was legal then isn't now. [popbottle, Jan 24 2014, last modified Apr 25 2017]

So, when they ask to move their stocks around, for example to another broker, what exactly happens?
-- MaxwellBuchanan, Jun 16 2008


Well, then you'd have to buy some real stock for them to move to the other broker, but this doesn't stop the business model from working. The only thing which might cause this to fail is when people applying for new day-trading accounts are rejected because they appear to know what they're talking about and obviously inept people are given top-spec accounts with generous margin facilities and options trading - this is the sort of thing which might arouse suspicion.
-- hippo, Jun 17 2008


The problem with this idea: stocks provide a net gain (i.e. gambler wins), whereas traditional gambling results in a net loss (i.e. bookie wins).
-- xaviergisz, Jun 17 2008


I think the missing link here is that all the money you take should also be invested in the stock market, but with more skill than your site users.

[+] for a great idea anyway.
-- vincevincevince, Jun 17 2008


[-] Already baked. They call that fraud.
-- Jscotty, Jun 17 2008


//stocks provide a net gain (i.e. gambler wins)//

Yes, but I think the idea specifically excludes those investors who are likely to hang on long enough to earn a dividend or a slice of real-world capital growth from underlying assets.

Still, this is really just an on-line version of the kind of race-course con that Damon Runyan used to write about.
-- pertinax, Jun 17 2008


Really close to a ponzi scheme.

Still, though, as a very very amateur investor and a guy that admits to knowing almost nothing about the market works, it'd be cool to have a somewhat milder introduction other than jumping in with both feet.
-- Noexit, Jun 17 2008


Just because technically its fraud doesn't mean its a not a good idea (for my benefit anyway). And assuming I dont screw it up and go bankrupt it won't hurt the investors either.
-- ServoMan314, Jun 17 2008


[maxwell] has already flagged the one problem, otherwise this is a pretty good idea.
-- wagster, Jun 18 2008


You could argue that investment banks already do this, except they don't, quite. Essentially this idea is to take a bet that when certain stupid people invest in a stock, the price of that stock will fall. Investment banks make investments on behalf of their clients and also trade with their own money. If they think the price of a stock will fall, the bank will make money by shorting it - that is, selling stock they don't have to the market, in order to buy it later (to balance their books) at a lower price. So, if an inept investor asks the bank to buy a small amount of this stock on their behalf, the bank's net position in the stock will still be 'short' - i.e. they won't have any of it. In reality this sort of mixing of the bank's and the bank's clients' trading positions is frowned upon, but it's similar in principle to this idea.
-- hippo, Jun 18 2008


It's not that far from the truth either. A typical stock broker's will operate on a three tier basis - the brokers themselves, wheeling and dealing, talking loudly as if they know what they are talking about, encouraging their list of clients to trade as much as possible (they earn commissions and fees on each trade)

Once trades are agreed in principal, the booking is done in terms of a bet e.g. 100,000 USD on Transiberian Expeditions Ltd - at this point, only a rough price will have been agreed with the client, and a 'ticket' written out and sent over to the traders.

In a large company, there will be a number of 'cross-trades' where one client is buying, and another happens to be selling the same thing. A sharp trader will settle these in-house, at the prices obtained from one of the markets, with no additional market involvement. They'll still charge all the usual brokerage, market and admin fees of course - so that on the 'client-side' of the trade, it looks like a traditional market-based trade.

In other cases, more than one client will buy or sell the same thing, allowing the trader to get a cheaper overall price (as with anything, large volumes buys you discounted prices) - the clients are of course charged full rates, allowing the trader to make a little 'margin', or 'spread'. It's this price which is finally quoted back to the client via the broker.

Finally, there is the settlement department who get an information feed from both the trading system, and confirmations of trades from the external trading parties. A big problem for settlement is failed trades where somewhere along the line, there's been a cock-up. When these happen, it's important to do right by the client, and present to them a view as though the thing has been bought as per the initial contract. This can be expensive if there has been a lot of movement in the price between the original trade date, and the point at which it's discovered not to have taken place. The brokerage is generally expected to foot the bill at this point - and can either choose to purchase the security at the higher price at the earliest practical opportunity, or sweat it out and hope the client closes their position asap - ideally, the client will know nothing of this - all the risk lying at the brokerage - just like in the idea, and [hippo]'s mention of 'shorting'.

So while it isn't generally accepted, this does already happen exactly as described.

That being said, regulatory bodies such as the FSA or the SEC would very likely frown upon a 'virtual' brokerage like this unless it was backed by very large sums of capital.

Of course, you could represent yourself as a bookie - but would have to conform to a different set of regulations - and still be seen to maintain a large 'float' in order to get you through the bad times.

Overall though, you'd probably make more money by acting as a traditional broker, where your revenue is generated on fees, commissions and 'spreads'.
-- zen_tom, Jun 18 2008


I don't see the point of this. A legitimate broker takes a fee on the deal and doesn't care whether the price goes up or down, that's the customer's problem. Given that, as zen_tom says, doing it the legal way you are guaranteed an income for doing next to nothing, what benefit would there be in not buying the stock as requested that outweighs the risk of being caught and imprisoned?

Also, the concept of dealing in stocks which don't actually exist is hardly new either. That's what stock derivatives and futures are. And when you get in a mess with them you get in a big mess.
-- DrBob, Jun 19 2008


The reason day traders lose money is the fees they pay, not because stocks somehow magically go down in the short term while going up in the long term. If you don't charge fees you will lose money as long as stocks go up on average. If you do charge fees, I agree with Dr. Bob.
-- Ford, Jun 19 2008


Already been done. They're called CFDs.
-- nemesis, Nov 04 2008


The problem with the idea is that your taking too much risk. A bet that you make with a bookie uses odds or a spread to make it an even bet. If you win you get paid, if you lose you lose the amount of the bet and an additional fee that's about a fifth of the bet called the "vig". Any bookie tries to get an equal amount of money on each side of the bet. The losers money pays the winners and the bookies profit is the vig.

The problem in this scenero is that there are no odds, too many stocks and the "vig", your commission is too small. You have no way to balance the winners against the losers and if you did your commission is too small to make any money unless your volume is very high. If the market has a good day and the majority of stocks go up you're in big trouble because a day-trader will cash out.

To reduce your risk you should aim not for people who put money in and take it out daily, but people who put it in and leave it in for a long time. You want the Bond market. You offer a bond the pays 10% and matures in 20 years with no early withdrawals. You give them a computer account that shows the interest piling up. 20 years gives you a long time to find a good hiding place.
-- theGem, Nov 06 2008


[-] on account of the fraud thing, as Jscotty pointed out. There are so many ways to get stupid people to give you their money, there's no need to mislead anyone.

// If [investment banks] think the price of a stock will fall, the bank will make money by shorting it... //

Adding to hippo's first paragraph comments, it's my understanding that trades like this made by large investment banks and hedge funds are sometimes automated by complex algorithms that do real-time market analysis. That is, not only are the decisions based on predictive models, but trades are actually executed by the software with no human interaction or judgment.

I came across this info in the software and technology world, so I have no idea of its actual prevalence in the investment/financial domain. But from a technology perspective, it's kind of cool and kind of scary at the same time.
-- Capt Skinny, Nov 06 2008



random, halfbakery